Professional Documents
Culture Documents
Learning Techniques
Hareesh Bahuleyan
Problem Statement
● Music genres are a way to classify
music based on rhythmic structure,
harmonic content and
instrumentation
● Automatically recognition
○ Organize digital libraries
○ Provide recommendations
Data
Google Audio Set
● 2.1 Million audio samples (of 10 seconds)
● 527 classes of sounds
● Selected 7 labels
Source: http://www.wildml.com/2015/11/understanding-convolutional-neural-networks-for-nlp/
VGG-16
● Transfer Learning
○ Weights of conv base are fixed
● Fine Tuning
○ Both conv base and feed-forward network are trainable
Feature Engineering
Approaches
Feature Extraction
7. Tempo
● Spectral Band-width
● Spectral Contrast
○ Divide spectrum into frequency bands
○ Maximum magnitude - Minimum magnitude in each band
● Spectral Roll-off
○ Frequency below which 85% of the total energy in the spectrum lies
● Chroma Features
○ 12-element feature vector
○ Indicates how much energy of each pitch class, {C, C#, D, D#, E, ..., B}
Results
Comparison of Models
● Metrics: Accuracy | F-score | AUC
Ensembling classifiers
is beneficial
Feature Importance Study
Keep only
most
important
top N
features
Many mis-classifications
for Rhythm blues, Pop
genre